Tuesday, Aug 30, 2016
BLOGPOST: The data is ‘Big’, the market is booming, ROI is excellent and the “things” landscape is heating up. There are a lot of platforms and services available which enables quick turnaround for deploying Big Data infrastructure or platforms. Companies want to invest more to figure out that whether the data they have stored over years in databases/generate daily can be monetized using data mining / analytics. Currently every company small or big wants to do “something” with ‘Big Data’, often projects are in a jeopardy as • the problem statements is not comprehended well enough • hypothesis for the ‘value’ generated is not evaluated enough • data by itself is not ‘Big’ enough • there are other unknown conditions In this post we will discuss the end-to-end problems that are comprehended in a data science/analytics projects both architecturally and as-is, we would also attempt to engineer a process/workflow which would help to determine the viability/need of a Big Data system given the problem statement.